Content uploaded by Peng Wang
Author content
All content in this area was uploaded by Peng Wang on May 10, 2016
Content may be subject to copyright.
Available via license: CC BY 4.0
Content may be subject to copyright.
The Dynamic Changes of DNA Methylation and Histone
Modifications of Salt Responsive Transcription Factor
Genes in Soybean
Yuguang Song, Dandan Ji, Shuo Li, Peng Wang, Qiang Li, Fengning Xiang*
The Key Laboratory of Plant Cell Engineering and Germplasm Innovation, School of Life Sciences, Shandong University, Jinan, Shandong, China
Abstract
Epigenetic modification contributes to the regulation of gene expression and plant development under salinity stress. Here
we describe the identification of 49 soybean transcription factors by microarray analysis as being inducible by salinity stress.
A semi-quantitative RT-PCR-based expression assay confirmed the salinity stress inducibility of 45 of these 49 transcription
factors, and showed that ten of them were up-regulated when seedlings were exposed to the demethylation agent 5-aza-2-
deoxycytidine. Salinity stress was shown to affect the methylation status of four of these ten transcription factors (one MYB,
one b-ZIP and two AP2/DREB family members) using a combination of bisulfite sequencing and DNA methylation-sensitive
DNA gel blot analysis. ChIP analysis indicated that the activation of three of the four DNA methylated transcription factors
was correlated with an increased level of histone H3K4 trimethylation and H3K9 acetylation, and/or a reduced level of H3K9
demethylation in various parts of the promoter or coding regions. Our results suggest a critical role for some transcription
factors’ activation/repression by DNA methylation and/or histone modifications in soybean tolerance to salinity stress.
Citation: Song Y, Ji D, Li S, Wang P, Li Q, et al. (2012) The Dynamic Changes of DNA Methylation and Histone Modifications of Salt Responsive Transcription
Factor Genes in Soybean. PLoS ONE 7(7): e41274. doi:10.1371/journal.pone.0041274
Editor: Xiaoyu Zhang, University of Georgia, United States of America
Received May 7, 2012; Accepted June 19, 2012; Published July 18, 2012
Copyright: ß2012 Song et al. This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits
unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Funding: This work was supported by the National Special Science Research Program of China (grant no. 2007CB948203) http://www.973.gov.cn/AreaMana.aspx;
the National Natural Science Foundation of China (grant nos. 30970243 and 30771116) http://www.nsfc.gov.cn; Excellent Youth Foundation of Shandong Province
of China (grant no. JQ200810) http://www.sdnsf.gov.cn/portal/; Science & Technology Plan of Shandong Province (grant 2009GG10002001) http://www.sdstc.gov.
cn/index.jsp and the Chinese Natural Education Ministry Doctor Station Foundation Fellowship (grant no. 913111006) http://std.nankai.edu.cn/kyjh-bsd/1.htm.
The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.
Competing Interests: The authors have declared that no competing interests exist.
* E-mail: xfn0990@sdu.edu.cn
Introduction
Soybean (Glycine max (L). Merr.) is an important source of
protein and oil in both the human and domestic animal diet. As
for most crop species, its productivity is significantly compromised
by soil salinity [1], but, like most plants, it has evolved a variety of
mechanisms to aid its survival under environmental stress. The
expression of many plant genes is altered by salinity stress; some of
these encode aspects of cellular metabolism and stress tolerance,
while others are regulatory in nature [1,2]. Transcription factors
(TFs), which belong to the latter class, have been classified into a
number of families on the basis of their sequence, and some
members of the MYB,NAC,b-ZIP and AP2-DREB families have
been shown to be intimately involved in the stress response
[3,4,5,6,7,8]. Such as, the heterologous expression of three
soybean MYB and three b-ZIP TFs in Arabidopsis thaliana improved
its response to salinity and freezing stress [9,10]. Similarly the
heterologous expression of GmDREB2 was able to enhance the
drought and salinity tolerance of tobacco [11], as did the over-
expression of either GmNAC11 or GmNAC20 for soybean [12].
Once a plant detects the onset of stress, TFs characteristically
respond by inducing the expression of a cascade of downstream
targets. However, their activation is in part also dependent on their
chromatin structure, which is largely determined by epigenetic
means [13,14,15,16]. Cytosine methylation within the promoter
sequence has been shown to underlie numerous cases of gene
down-regulation or silencing [17,18,19,20,21]. DNA methylation
in the plant genome mostly at CG dinucleotides and CNG
trinucleotides, but also at an asymmetrical sequence contexts CNN
(N is any nucleotide but G) [22,23,24]. The N terminus of the
histone molecule can be acetylated, phosphorylated, methylated,
ubiquitinated or ribosylated [25]. The presence of the trimethy-
lated form of histone H3K4 and of the acetylated form of H3K9 in
the promoter region have been frequently associated with
transcriptional activation, while that of the dimethylated form of
H3K9 represses it [26,27,28]. Sometimes, H3K9 methylation can
trigger cytosine methylation in both Neurospora crassa [29] and A.
thaliana [30], while cytosine methylation at the CNG trinucleotide
appears to be partially dependent on the activity of a histone
methyltransferase [29,31,32].
A number of examples where epigenetic modification has
contributed to the regulation of gene expression during periods of
environmental stress have been presented. In particular, the low
temperature induced expression of the maize gene ZmMI1 has
been correlated with a reduction of DNA methylation in its
nucleosome core [33]. In tobacco, several stress agents are known
to promote demethylation in the NtGPDL coding sequence, leading
to alterations in its level of expression [34]. The submergence of
rice seedlings reduces histone H3K4 trimethylation and acetyla-
tion in genes encoding both alcohol dehydrogenase and pyruvate
decarboxylase, leading to their up-regulation [35]. In A. thaliana,
the drought-induced expression of a number of stress-responsive
PLoS ONE | www.plosone.org 1 July 2012 | Volume 7 | Issue 7 | e41274
genes has been associated with an increase in H3K4 trimethyla-
tion and H3K9 acetylation [36]. Here, we set out to document the
induction by salinity stress of DNA methylation and histone
modification in a number of salinity responsive soybean TFs, and
to identify what relationship there is, if any, between the
expression of a TF and the epigenetic status of its promoter
sequence. In addition, since to date no systematic attempt has
been made to investigate the dynamics and reversibility of both
DNA and histone modification over the course of a stress episode,
we have explored this feature focusing on four salinity stress
inducible soybean TFs.
Materials and Methods
Exposure of soybean seedlings to salinity stress and 5-
ADC treatment
Seedlings of the soybean cultivar Williams 82 were grown in
vermiculite under a 16h photoperiod at 25uC for 14 days before
being exposed to stress treatment. Once the seedlings had been
removed from the vermiculite and their roots rinsed in water, they
were then treated with either 150 mM NaCl for 1h, 3h, 6h, 12h or
24h, or with 50 mM 5-aza-2-deoxycytidine (5-ADC) for 12h, 24h,
48h or 72h. RNA and DNA was extracted from snap-frozen plants
both before the stress treatment had begun and then at each time
interval. Mock treatments (ddH
2
O only) were included as a
control. RNA was prepared from 0.2 g plant material using the
TRIzol (Invitrogen) reagent, following the manufacturer’s proto-
col, and DNA was extracted from 1 g plant material using a
DNeasy Plant Mini kit (Qiagen).
Microarray analysis
RNA were isolated from the mock (M0, M1, M3, M6, M12,
M24) and salinity treated (S0, S1, S3, S6, S12, S24) seedlings.
0.5 mg RNA that extracted from each time point of the mock and
salinity-stressed seedlings were mixed respectively to obtain the
mock and salinity-stressed RNA pools, and then they were used to
synthesize the cDNA. The cDNA was labeled with biotin, and
then hybridized to an Affymetrix soybean Genome Array
according to the manufacturer’s instructions (15h in a rotating
hybridization oven set at 45uC and 60 rpm). After the hybridiza-
tion, the microarrays were scanned using a GeneChipHScanner
3000 (Affymetrix, P/N 00-00212). Then the scaling factor,
background, noise, and percentage presence were calculated
according to the Affymetrix Data Mining Tool protocols
(Affymetrix). All resulting datasets were filtered using the absolute
call metric (present or absent) implemented within Microsoft
Access (Microsoft Corporation, Redmond, WA) and the micro-
array data were processed in an R (v2.7.0) environment, using the
LIMMA package [37]. Quantile normalization was performed. A
single repeat microarray analysis for each group was performed.
Figure 1. The expression of the 49 TFs in mock-stressed and salinity-stressed seedlings. (a) GmAP2-DREBs, (b) GmMYBs, (c) GmNACs and
(d) Gmb-ZIPs. M0-M24 refer to seedlings exposed to just ddH
2
O for, respectively, 0h, 1h, 3h, 6h, 12h and 24h; S0–S24 refer seedlings exposed to
150 mM NaCl for 0h, 1h, 3h, 6h, 12h and 24h, respectively. Each gene-specific region was amplified by RT–PCR using the gene-specific primers
(Table S2). The TUBULIN gene (Genbank accession AY907703) was used as an internal control. The experiment was repeated three times with similar
result.
doi:10.1371/journal.pone.0041274.g001
Epigenetic Modifications Affect Salt Response
PLoS ONE | www.plosone.org 2 July 2012 | Volume 7 | Issue 7 | e41274
Transcript level analysis
Semi-quantitative RT-PCR (sqRT-PCR) and quantitative real-
time RT-PCR (qRT-PCR) were employed to quantify transcript
levels more precisely. RNA was extracted from 0.2 g of seedling
material ground in liquid nitrogen by the addition of 1 ml
TRIZOL reagent (Invitrogen) and treated with RNase-free DNase
I. A 3 mg aliquot of total RNA was used to generate the first cDNA
strand with the SuperScript First-Strand Synthesis System
(Invitrogen) according to the manufactuter’s instructions. A 1 ml
aliquot of this cDNA was used as the template for a 22–34 cycle
sqRT-PCR, where the cycling regime was 94uC/30 s, 55uC/30 s,
72uC/30 s. A fragment of the soybean TUBULIN gene (Genbank
accession AY907703) was used as a reference. Primer sequences
are given in Table S2. Each 15 ml qRT-PCR contained 7.5 ml
Maxima SYBR Green qPCR Master mix buffer (Roche), 0.5 ml
10 mM specific primers, 1.5 ml of a 1:10 dilution of cDNA and
5.5 ml ddH
2
O. The cycling regime consisted of a denaturation step
(95uC/3 min) followed by 18–35 cycles of 95uC/30 s, 60uC/15 s,
72uC/15 s, and a fragment of the soybean TUBULIN gene
(GenBank accession AY907703) was used as a internal control.
Primer sequences are given in Table S2. The relative expression
level of the target sequence was determined using the 2
2DDCt
method [38]. Each estimate was derived from the mean of three
independent biological replicates.
Bisulfite DNA sequencing
A2mg DNA aliquot was dissolved in 50 ml ddH
2
O and
denatured by adding 5.5 ml 3 M NaOH and incubating for
30 min at 42uC. Thereafter, 510 ml 2.3 M sodium bisulfite
(pH 5.0), 30 ml 10 mM hydroquinone and 65 ml ddH
2
O were
added, and the solution overlaid with mineral oil and held for 16h
at 55uC. The DNA was recovered using a WizardHDNA Clean-
Up System kit (Promega A7280), and a 90 ml aliquot treated with
10 ml 3 M NaOH for 15 min at 37uC, then neutralized by adding
70 ml 10 M ammonium acetate. Finally, the DNA was precipitat-
ed by adding 400 ml ethanol and 10 ml glycogen, and re-suspended
in 50 ml ddH
2
O to provide the template for a series of PCRs based
on the gene-specific primers listed in Table S2. A fragment of
Glyma20g32730 featuring many CG, CNG and CNN sites was
amplified from genomic DNA, then inserted into pMD18-T vector
and transferred into Dm- E. coli strain JM110. The plasmid was
released from the bacterial cells by the plasmid extraction kit
(TianGen. Cat. DP103-03) and treated with bisulfite in parallel
with the soybean genomic DNA as a control to monitor the
transformation efficiency of unmethylated cytosine to thymine.
The subsequent PCR consisted of 34–37 cycles of 94uC/30 s,
55uC/30 s, and 72uC/40 s. The resulting amplicons were purified
with a WizardHDNA Clean-Up System kit, ligated into the
pMD18-T vector (TaKaRa) and transferred into E. coli for
sequencing. Ten clones from each amplicon were sequenced.
Figure 2. Expression of 45 salinity inducible TFs in seedlings exposed to 5-ADC treatment. (a) GmAP2-DREBs, (b) GmMYBs, (c). GmNACs.
and (d) Gmb-ZIPs. M0-M72 refers to seedlings treated with water only for, respectively 0h, 12h, 24h, 48h and 72h, while A0-A72 refer to seedlings
exposed to 50 mM 5-ADC for 0h, 12h, 24h, 48h and 72h, respectively. Each gene-specific region was amplified by RT–PCR using the gene-specific
primers (Table S2). The TUBULIN gene (Genbank accession AY907703) was used as an internal control. The experiment was repeated three times with
similar result.
doi:10.1371/journal.pone.0041274.g002
Epigenetic Modifications Affect Salt Response
PLoS ONE | www.plosone.org 3 July 2012 | Volume 7 | Issue 7 | e41274
Figure 3. Methylation status of the promoter region of four salinity-responsive TFs in untreated (S0) and salinity-stressed (S1–S24)
seedlings (S1: 1h, S3: 3h, S6: 6h, S12: 12h, S24: 24h). (a) The black and white boxes indicate, respectively, exon and untranslated regions. The
short bars annotated with ‘‘I, II, III’’, ‘‘a’’ or b’’ indicate, respectively the sequences subjected to ChIP analysis, genomic bisulfite sequencing and those
used as probes for Southern blotting. The long vertical bars marked ‘‘c’’ display the distribution of CG dinucleotides (marked with red vertical lines),
and CNG (blue vertical lines) and CNN (black vertical lines) trinucleotides. The red vertical lines marked with a rectangular indicate CCGG sites
analyzed by Southern blotting. The thick black vertical lines represent the proportion of methylated cytosine. Ten positive clones from each gene’s
amplicon were sequenced. The data reflect the outcome of three independent experiments, and error bars represent standard error (SD). (b) The
efficiency of the bisulfite treatment to transform unmethylated cytosine to thymine. A fragment of Glyma20g32730 with numerous cytosines was
cloned into Dm- E. coli cells and the plasmid was treated with bisulfite in parallel with the soybean genomic DNA. All clones processed showed a
transformation rate .99.7%. (c) Methylation-sensitive DNA gel blot analysis of non-stressed (S0) and salinity-stressed seedlings (S1–S24). Genomic
DNA was digested to generate large fragments, then with one or other of the schizomers HpaII or MspI. Hybridization probes indicated. A DNA
fragment amplified from the probe sequence was used as a positive control (+), and ddH
2
O was used as a negative control (2).
doi:10.1371/journal.pone.0041274.g003
Epigenetic Modifications Affect Salt Response
PLoS ONE | www.plosone.org 4 July 2012 | Volume 7 | Issue 7 | e41274
The process was repeated three times using biologically indepen-
dent samples.
Chromatin immunoprecipitation (ChIP) assay
The ChIP protocol was modified from that of Johnson et al.
(2002). Briefly, 1 g of plant tissue was fixed by immersion in 1% v/
v formaldehyde under vacuum for 10 min. The extracted DNA/
protein complex was then sheared by sonication to a size range of
,100–1000 bp. After centrifugation, the complex was immuno-
precipitated by challenging with H3K9ac, H3K4me3 and
H3K9me2 antibodies (Millipore cat. 07–392, 07–473 and 05–
768R) at a titer of 1:100. The residual protein was degraded by the
addition of 10 ml (20 mg/ml) proteinase K, followed by a phenol/
chloroform extraction. A 2 ml aliquot of the final solution was used
as a template for qRT-PCR analysis as described above. A 1000x
diluted input DNA (Input) obtained from 500 ml of extract was
purified in parallel with the immunoprecipitated samples as a
control, and ChIP reactions were also performed in the absence of
antibody (No AB) to detect the occurrence of any non-specific
binding. Relative levels of H3K9 acetylation, H3K9 dimethylation
and H3K4 trimethylation were normalized to an internal control
(GenBank accession AY907703). The sequences of all PCR
primers used are given in Table S2. The mean and standard
deviation are shown for three independent ChIP experiments and
the significance of differences between means assessed with a ttest.
Methylation-sensitive Southern blot analysis
Genomic DNA (100 mg) extracted from both non-stressed and
salinity-stressed seedlings was treated for at least 6h with 100 U of
the appropriate restriction enzymes (EcoRV and NdeI for
Glyma11g02400;SacI for Glyma16g27950;BglII for Glyma20g30840;
SacI and BglII for Glyma08g41450) (TaKaRa) to generate large
fragments containing the target sequences. The digested DNA was
extracted by phenol/chloroform and divided into two equal
aliquots, one of which was treated with HpaII and the other with
its schizomer MspI [39]. The digested DNA was re-extracted,
electrophoretically separated through an 0.8% agarose gel and
transferred onto a Hybond N
+
membrane (Amersham). Probes for
each gene were designed to detect the methylation status within
the target sequence that analysed by genomic bisulfite sequencing.
About 3 mg of probe DNA was labeled using a DIG-High Prime
kit (Roche), and the subsequent hybridization and detection
procedure was performed using a DIG High Prime DNA Labeling
and Detection Starter Kit I (Roche), according to the manufac-
turer’s instructions. The positive control consisted of a 100x
diluted DNA fragment that amplified from the genomic DNA in
the same regions that prepared for probes, while the negative
control was ddH
2
O.
Results
The identification of salinity stress responsive TFs in
soybean
A set of differentially expressed genes were identified by
comparing the soybean Affymetrix microarray profiles generated
by probing with RNA extracted from salinity-stressed and non-
stressed plants. We mainly focused on the four groups of AP2/
EREB, bZIP, NAC and MYB transcription factors that have been
verified for salt stress in Arabidopsis or other plants. Of the 1,335
MYB, NAC, AP2/DREB and b-ZIP TFs represented on the
microarray, 49 appeared to be up-regulated (fold change of
hybridization signal .2, p,0.01) by salinity stress. These consisted
of 15 (of 448) GmMYBs, 9 (of 226) GmNACs, 16 (of 426) GmAP2/
DREBs and 9 (of 235) Gmb-ZIPs (Table S1). When the expression
of these 49 TFs was assayed by sqRT-PCR in both mock-stressed
(M0–M24) and salinity-stressed soybean seedlings (S1–S24) with
gene-specific primers (Table S2), 14 of the GmMYBs, 8 of the
GmNACs, 15 of the GmAP2/DREBs and 8 of the Gmb-ZIPs were
confirmed to be markedly induced by salinity stress (Figure 1).
Expression pattern analysis indicated that 19 of them were
strongly induced at a relatively early stage of exposure to salinity
(1–3h), while the others were induced somewhat later (6–24h).
Expression of the salinity induced TFs in the presence of
5-ADC
The expression of the 45 salinity-induced TFs was then
monitored in the mock treated (M0–M72) and the seedlings that
exposed to 5-ADC for various periods (A0–A72). As a result, ten of
the them showed higher levels of expression in treated (M0–M72)
than in mock-treated (A0–A72) seedlings; these ten TFs consisted
Figure 4. DNA methylation patterns in
Glyma11g02400
,
Glyma16g27950
,
Glyma08g41450
and
Glyma20g30840
in non-stressed (S0) and
salinity-stressed (S1 S24) seedlings. The left axis shows the percentage of methylated cytosines at each site (present as CG, CNG and CNN). The
data represent the mean of three biological replicates. Error bars represent standard errors.
doi:10.1371/journal.pone.0041274.g004
Epigenetic Modifications Affect Salt Response
PLoS ONE | www.plosone.org 5 July 2012 | Volume 7 | Issue 7 | e41274
of four GmMYBs(Glyma11g02400,Glyma07g30860,Glyma12g34650,
Glyma15g07230), one GmNAC (Glyma15g08480), four GmAP2/
DREBs(Glyma20g32730,Glyma20g30840,Glyma16g27950,Gly-
ma10g00980) and one Gmb-ZIP genes (Glyma8g41450) (Figure 2).
The expression level of nine of these TFs was very low for the first
12h of exposure, but thereafter rose substantially; the exception
was the Gmb-ZIP Glyma08g41450, the expression of which was
induced somewhat earlier.
Figure 5. Promoter methylation status in four salinity-responsive TFs in non-treated (A0) and 5-ADC treated seedlings (A12–A72).
For Figure legend please refer to Figure 3 legend.
doi:10.1371/journal.pone.0041274.g005
Epigenetic Modifications Affect Salt Response
PLoS ONE | www.plosone.org 6 July 2012 | Volume 7 | Issue 7 | e41274
Figure 6. Expression, DNA methylation and histone modification status of
Glyma11g02400, Glyma16g27950, Glyma08g41450
and
Glyma20g30840
in none treated (S0) and salinity-stressed (S1–S24) seedlings. (a) Relative H3K9 demethylation, acetylation and H3K4
Epigenetic Modifications Affect Salt Response
PLoS ONE | www.plosone.org 7 July 2012 | Volume 7 | Issue 7 | e41274
Methylation status as affected by salinity stress
To investigate the DNA methylation status of above candidate
genes under salinity stress, the sequence corresponding to the
translation start codon and the promoter region of the ten TFs was
subjected to bisulfite sequencing. First, the efficiency of the sodium
bisulfite treatment to convert cytosine to thymine was estimated.
The efficiency of the sodium bisulfite treatment to convert cytosine
to thymine in the cytosine rich segment of Glyma20g32730 was
estimated to be 99.7%. Bisulfite sequencing result indicated that
the Glyma11g02400, Glyma08g41450, Glyma16g27950 and Gly-
ma20g30840 promoters all appeared to be differentially methylated
by the imposition of salinity stress (Figure 3a, b, Figure S1), but
those of the other six genes were largely non-methylated
(Figure S2). In the Glyma11g02400 promoter from position 2518
to 2274, most of the cytosines were demethylated following
exposure to salinity stress for 1–24h (Figure 3a). In the immediate
downstream region of the Glyma16g27950 transcription start codon
(+24 to +233), about 35% of the cytosines were methylated both
before the salinity stress was imposed and for the first three hours
of stress, but thereafter only few methylated cytosines remained
(Figure 3a). In the Glyma20g30840 promoter region 1 (287 to
+163), 51% of the cytosines were methylated prior to exposure to
salinity stress, but this proportion fell to 27% after 1h, 12% after
3h, and to even lower levels as the stress was prolonged further
(Figure 3a); meanwhile in region 2 of the same promoter (2163 to
2405), 42% of the cytosines were methylated at the start of the
stress period, and this proportion hardly altered thereafter
(Figure 3a). In the Glyma08g41450 region immediately downstream
of the transcription start codon (+24 to +233), 35% of the cytosines
were methylated prior to the imposition of stress, and the same
proportion was maintained throughout (Figure 3a).
DNA methylation-sensitive Southern blotting was applied to
verify these observations. The restriction fragments including
Glyma11g02400,Glyma16g27950 and Glyma20g30840 were more
readily digested by HpaII after 6h of salinity stress than the same
samples obtained from non-stressed seedlings, which consistent
with a reduction in global cytosine methylation caused by salinity
stress analyzed by bisulfite sequencing (Figure 3a, c). The sequence
surrounding Glyma08g41450 was not digestible by HpaII, but
several small restriction fragments were generated by MspI
digestion, suggesting that CCGG sites in the region of this TF
were hypermethylated in both non-stressed and stressed seedlings
(Figure 3c). Thus the Southern blotting outcomes were in general
consistent with the bisulfite sequence data.
The analysis of DNA methylation pattern of them indicated that
methylation was affecting either CG dinucleotides or CNG/CNN
trinucleotides under salinity stressed process (Figure 4). In the
Glyma11g02400 promoter, 98% of the CG’s, 60% of the CNG’s
and 6% of the CNN’s were methylated in the non-stressed
seedlings, but almost all the CG’s, CNG’s and CNN’s were
demethylated in plants exposed to salinity stress for more than 6h
(Figure 4). For Glyma16g27950, some 80% of the CG’s, 72% of the
CNG’s and 4% of the CNN’s were methylated in non-stressed
seedlings and those during the early phase (1–3h) of the salinity
treatment, but by 6h a significant fall in CG and CNG methylation
was observed (Figure 4). Within region 1 of the Glyma20g30840
promoter, 95% of the CG’s, along with 40% of the CNG’s and 4%
of the CNN’s, were methylated at 3h after the imposition of stress,
but by 6h, a marked reduction in CG and CNG methylation had
occurred. CG, CNG and CNN methylation in the Glyma08g41450
promoter was unaffected by salinity stress (Figure 4). Clearly, DNA
methylation in Glyma11g02400, Glyma16g27950 and Gly-
ma20g30840 (region 1) varied over the period of the salinity stress
episode.
Methylation status of Glyma11g02400, Glyma08g41450,
Glyma16g27950 and Glyma20g30840 as affected by the
presence of 5-ADC
To identify whether the up-regulation of these four genes were
related with cytosine demethylation under 5-ADC treatment, the
effect on the DNA methylation status of the four responsive TFs in
plants treated with 5-ADC was analyzed using genomic bisulfite
sequencing. The relevant test for the transformation efficiency of
unmethylated cytosine to thymine using Glyma20g32730 is
illustrated. As a result, all four TFs were hypermethylated in
non-treated seedlings; after a 24h exposure to 5-ADC, some
evidence of demethylation was obtained, but from 48h onwards it
was clear that a substantial level of demethylation had occurred
(Figure 5a, b). This observation was supported by a DNA
methylation-sensitive DNA gel blot (Figure 5c). All of the four
TFs showed an increased digestion with HpaII after salinity stress
for more than 48h, suggesting a reduction of cytosine methylation
in 5-ADC stressed seedlings (Figure 5c).
Histone modification of hypermethylated genes induced
by salinity stress
The histone content (H3K4me3, H3K9ace and the inactive
H3K9me2) of the four TFs (Glyma11g02400,Glyma08g41450,
Glyma16g27950 and Glyma20g30840) which responded to salinity
stress by altering their methylation status was then examined,
using a combination of ChIP and qRT-PCR (Figure 6a). An
unmethylated gene Glyma20g32730 was also analysed in parralle
with them as a control (Figure S3). When Glyma11g02400 was
induced by salinity stress, a significant increase in H3K4me3
(regions I, II and III) and a decrease in H3K9me2 (regions I and
II) was observed, but the H3K9ace sites remained unmodified
(Figure 6a). Within Glyma20g30840 (regions II and III) and
Glyma08g41450 (region III), a high level of H3K9me2 and a low
level of H3K4me3 and H3K9ac was present in both non-stressed
seedlings and those sampled during the early phase (1–3h) of
salinity stress; at later time points (6–24h), a significant decrease in
H3K9me2 and increase of H3K4me3 and H3K9ac content was
observed (Figure 6a). A similar H3K4me3, H3K9ac or H3K9me2
signal was detected in all three regions of the Glyma16g27950
promoter (Figure 6a). Thus, like the DNA methylation, histone
modification was also subject to dynamic change during the course
of the salinity stress episode.
Changes of epigenetic modification in regulating the TFs
expression during salinity stress
The expression of Glyma11g02400 was low in non-stressed
seedlings, while its promoter was hypermethylated and was highly
enriched for H3K9me2 and depleted for H3K4me3 (Figure 6a, b,
trimethylation content (ChIP assay). A 1:1,000 dilution of input DNA (Input) served as a control for PCR amplifications and the ChIP reactions carried
out in the absence of antibody (N0 AB). Relative H3K9 acetylation, H3K9 dimethylation and H3K4 trimethylation were determined by qRT-PCR and
normalized to an internal control TUBULIN gene (Genbank accession AY907703). Data represent the mean of three biological replicates. Asterisks
indicate means differing significantly from the S0 situation. Error bars represent standard errors. *P,0.05, **P,0.01. (b) Gene expression (qRT-PCR)
profiles. (c) Cytosine methylation level (bisulfite sequencing).
doi:10.1371/journal.pone.0041274.g006
Epigenetic Modifications Affect Salt Response
PLoS ONE | www.plosone.org 8 July 2012 | Volume 7 | Issue 7 | e41274
c). When the seedlings were exposed to salinity stress, its
expression rose rapidly, while its promoter became gradually
demethylated, the level of H3K9me2 fell and that of H3K4me3
increased (Figure 6a, b, c). Glyma20g30840 behaved similarly, but
the establishment of H3K4me3, DNA demethylation and gene
expression were not contemporaneous. Demethylation was
noticeable within 1h of the imposition of salinity stress, but the
establishment of H3K4me3 did not occur until 3h and the up-
regulation of expression only at 6h (Figure 6a, b, c). Similarly,
Glyma08g41450 was up-regulated by 12h after the start of the stress
episode, but its promoter was hypermethylated throughout.
Between 6h and 24h after the stress had begun, the level of
H3K9me2 fell and that of H3K4me3 and H3K9ac rose (Figure 6a,
b, c). Glyma16g27950 expression was repressed in non-stressed
seedlings and during the early phase of the stress episode, when its
promoter was hypermethylated. Later the TF was gradually
induced and its promoter demethylated, while the level of
enrichment of H3K9me2, H3K4me3 and H3K9ac did not
change. A correlation analysis of their expression, methylation
levels and histone modifications indicated that up-regulation of
Glyma11g02400 was associated with a decreased level of DNA
methylation (r = 20.89), H3K9me2 (r = 20.965, on average of
regions II and III) and an increased level of H3K4me3 (r =0.7, on
average of regions I and II) (Figure 6; Table S3). The expression of
Glyma20g30840 correlated negatively with DNA methylaion
(r = 20.78), H3K9me2 (r = 20.93 on average of regions II and
III) and positively with H3K9ac (r = 0.96) and H3K4me3 (r = 0.93
on average) (Figure 6; Table S3). Up-regulation of Glyma16g27950
just correlated negatively with DNA methylation (r = 20.97), did
not with histone modifications, while the up-regulation of
Glyma08g41450 correlated negatively with H3K9me2 (r = 20.67)
and positively with H3K9ac (r = 0.84) and H3K4me3 (r = 0.78),
did not with DNA methylation during salinity stress (Figure 6;
Table S3). Therefore, the histone accumulation/depletion and/or
cytosine methylation appears to underlie the activation by salinity
stress of these four TFs.
Discussion
TF transcription can be influenced by both DNA
methylation and/or histone modification in a region
specific manner
The regulation of genes via cytosine methylation and histone
modification is a well recognized component of the plant stress
response [14,34,40,41]. Both these epigenetic modifications are
region-specific, and can be dynamic over time [42,43,44,45]. Here
we have identified a set of ten salinity-induced, cytosine
methylation-dependent TFs, three of which displayed the expected
relationship between promoter cytosine methylation and gene
expression during the salinity stress process, but one of which
(Glyma08g41450) remained up-regulated even though it was in a
highly methylated state (Figure 1, Figure 3a). A similar unexpected
relationship has been noted for an embryogenesis-related gene in
carrot [42]. Many genes are expressed despite their promoter
region being highly methylated, so it seems probable that for
Glyma08g41450, its up-regulation in plants exposed to salinity stress
is independent of DNA methylation. One region of the
Glyma20g30840 promoter was hypermethylated in both stressed
and non-stressed plants, while its neighbouring region responded
to the stress by a reduction in methylation (Figure 3a). This
behaviour provides an example of region-specific regulation of
methylation. The gradual up-regulation of Glyma11g02400 in
salinity-stressed seedlings was accompanied by a decrease in CG,
CNG and CNN methylation (Figure 4), while the rapid up-
regulation of Glyma20g30840 and Glyma16g27950 was accompa-
nied by a decrease in only CG and CNG methylation (Figure 4),
suggesting heterogeneity in the genome for gene expression
regulation via DNA methylation.
Histone modification provides a second major mechanism of
epigenetic control over gene expression [46]. In a range of A.
thaliana stress-responsive genes, salinity stress has been shown to
increase trimethylation at H3K4 and decrease H3K9 demethyl-
ation [28]. The up-regulation of Glyma20g30840 and Gly-
ma08g41450 during the course of the salinity stress episode may
have been achieved by the depletion of H3K9me2 and the
enrichment of H3K4me3 and H3K9ac in regions II and III of
Glyma20g30840 and in region III of Glyma08g41450 (Figure 6a, b).
The up-regulation of Glyma11g02400 may have been brought
about by the depletion of H3K9me2 and the enrichment of
H3K4me3 in regions I, II and III (Figure 6a, b). Thus, as for DNA
methylation, the effect of salinity stress on histone modification
appears to be heterogeneous across the genome.
Transcriptional activation, DNA methylation and histone
modification are not simultaneous events
There was a distinct time lag between transcriptional activation,
DNA methylation and/or histone modification of Glyma11g02400,
Glyma20g30840 and Glyma08g41450. Thus, most of the cytosine
content of the Glyma11g02400 promoter was demethylated very
soon (within 1h) of the imposition of stress, but there was no
evidence for the TF’s activation before 3h (Figure 6b, c). Similarly,
Glyma20g30840 was progressively demethylated over the full 24h
of the stress episode, but its up-regulation was complete within 6h
(Figure 6b, c). In Glyma11g02400,Glyma20g30840 and
Glym08g41450, the H3K4me3 content was already increasing by
1h, but the up-regulation of TF expression occurred substantially
later (Figure 6a, b). Similar time lags have been noted for the
build-up of H3K4me3 within the coding regions of the A. thaliana
drought-related genes RD29A and RAP2.4 during drought stress
[36]. Furthern more, the DNA demethylation of Glyma20g30840
was prior to the erasure or establishment of H3K9me2, H3K4me3
and H3K9ac (Figure 6a, c) and all the three genes (Gly-
ma11g02400, Glyma20g30840 and Glym 08g41450) show an ealier
establishment of H3K4me3 than H3K9me2 and H3K9ac
(Figure 6a). Suggesting that there were also a time lage between
the DNA demethylation and establishment of histone modifaci-
tions.
The interplay between DNA methylation and histone
modification in the context of patterns of gene
expression
In A. thaliana, it has been demonstrated that the loss of CG
methylation in met1 plants has a large effect on H3K9me2 content,
leading to the idea that cytosine methylation can influence H3K9
modification [17,47,48,49]. H3K9me2, mediated by KYP/
SUVH4 and SUVH2, is also known to direct non-CG methylation
[39,50]. In the ‘two-step’ hypothesis for the regulation of
transcription, CG methylation directs H3K9 methylation and
H3K9 methylation recruits non-CG methylation [51]. Moreover,
hypermethylation of the stress inducible genes in Arabidopsis
correlated with the enrichment of H3K9me2 and depletion of
H3K9ac histones under salt stress conditions [16]. In this study,
the behaviour of Glyma11g02400 and Glyma20g30840 was consis-
tent with this model; as CG was progressively demethylated in
these TFs, the content of H3K9me2, H3K4me3 and/or H3K9ac
rose and meanwhile a lower level of non-CG methylation was
observed (Figure 6a, Figure 4). Note, however, that for
Epigenetic Modifications Affect Salt Response
PLoS ONE | www.plosone.org 9 July 2012 | Volume 7 | Issue 7 | e41274
Glyma16g27950, while salinity stress led to a noticeable demeth-
ylation of the DNA, it had little effect on the level of histone
modification (Figure 6a, c). The Glyma08g41450 promoter was
hypermethylated throughout the salinity stress period, while
H3K9me2 was depleted and H3K4me3 and H3K9ac accumulat-
ed (Figure 6a, c). For these two TFs, therefore, the evidence is that
DNA methylation had no influence on histone modification,
consistent with the behaviour of the A. thaliana genes TOUSLED
and RPA2 [52,53].
Supporting Information
Figure S1 Monitoring the non-methylated cytosine to
thymine transformation efficiency of the bisulfite se-
quencing procedure. A fragment of Glyma20g32730 with
numerous cytosines was cloned into Dm- E. coli cells and the
plasmid was treated with bisulfite. Nine positive clones of this gene
from the bisulfite treated DNA were sequenced and compared
with the untreated DNA sequence.
(PDF)
Figure S2 Genomic bisulphite sequencing of six un-
methylated genes in plants not exposed to salinity (S0)
and those stressed for 1–24 h (S1–S24). Ten clones of each
gene were sequenced. The short thick bars indicate the regions
subjected to sequencing, and the long vertical ones show the
distribution of CG dinucleotides (black circles), and CNG
(triangles) and CNN (unmarked) trinucleotides. The black vertical
line represents the proportion of methylated sites. Data represent
the mean of three biological replicates.
(PDF)
Figure S3 ChIP analysis to assess the unmethylated
gene Glyma20g32730’s H3K9me2, H3K9ac and
H3K4me3 content in plants challenged with salinity.
The short bars marked ‘‘a’’ indicate regions subjected to genomic
bisulfite sequencing (+1to+393); ‘‘I, II and III’’ indicate the
regions subjected to ChIP analysis.
(PDF)
Table S1 The set of TFs identified by microarray
analysis to be inducible by salinity stress.
(DOC)
Table S2 Sequences of primers employed in this
research.
(DOC)
Table S3 Correlation analysis among methylation lev-
els, gene expression and histone modifications of the
four TFs during salinity stress. NC: no correlation;
*P
,
0.05, **P
,
0.01.
(DOC)
Acknowledgments
We thank Dr. Robert Koebner (UK) for English editing of the manuscript.
Author Contributions
Conceived and designed the experiments: FX. Performed the experiments:
YS DJ SL PW QL. Analyzed the data: YS. Contributed reagents/
materials/analysis tools: FX. Wrote the paper: YS.
References
1. Phang TH, Shao G, Lam HM (2008) Salt tolerance in Soybean. J Integr Plant
Biol 50: 1196–1212.
2. Cheong YH, Moon BC, Kim JK, Kim CY, Kim MC, et al. (2003) BWMK1, a
rice itogen-activated protein kinase, locates in the nucleus and mediates
pathogenesis-related gene expression by activation of a transcription factor.
Plant Physiol 132: 1961–1972.
3. Choi HI, Hong JH, Ha JO, Kang JY, Kim SY (2000) ABFs, a family of ABA-
responsive element binding factors. J Biol Chem 275: 1723–1730.
4. Uno Y, Furihata T, Abe H, Yoshida R, Shinozaki K, et al. (2000) Arabidopsis
basic leucine zipper transcription factors involved in an abscisic acid-dependent
signal transduction pathway under drought and high-salinity conditions. Proc
Natl Acad Sci U S A 97: 11632–11637.
5. Sakuma Y, Liu Q, Dubouzet JG, Abe H, Shinozaki K, et al. (2002) DNA-
binding specificity of the ERF/AP2 domain of Arabidopsis DREBs, transcrip-
tion factors involved in dehydration- and cold inducible gene expression.
Biochem Biophys Res Commun 290: 998–1009.
6. Abe H, Urao T, Ito T, Seki M, Shinozaki K, Yamaguchi-Shinozaki K (2003)
Arabidopsis AtMYC2 (b-HLH) and AtMYB2 (MYB) function as transcriptional
activators in abscisic acid signalling. The Plant Cell 15: 63–78.
7. Chen M, Xu Z, Xia L, Li L, Cheng X, et al. (2009) Cold-induced modulation
and functional analyses of the DRE-binding transcription factor gene,
GmDREB3, in soybean (Glycine max L.). J Exp Bot 60: 121–35.
8. Bu QY, Jiang HL, Li CB, Zhai QZ, Zhang J, et al. (2008) Role of the
Arabidopsis thaliana NAC transcription factors ANAC019 and ANAC055 in
regulating jasmonic acid-signaled defense responses. Cell Res 18: 756–767.
9. Liao Y, Zou HF, Wei W, Hao YJ, Tian AG, et al. (2008a) Soybean GmbZIP44,
GmbZIP62 and GmbZIP78 genes function as negative regulator of ABA
signaling and confer salt and freezing tolerance in transgenic Arabidopsis. Planta
228: 225–240.
10. Liao Y, Zou HF, Wang HW, Zhang WK, Ma B, Z, et al. (2008b) Soybean
GmMYB76, GmMYB92, and GmMYB177 genes confer stress tolerance in
transgenic Arabidopsis plants. Cell Research 18: 1047–1060.
11. Chen M, Wang QY, Cheng XG., Xu ZS, Li LC, et al. (2007) GmDREB2, a
soybean DRE-binding transcription factor, conferred drought and high-salt
tolerance in transgenic plants. Biochem Biophys Res Commun 353: 299–305.
12. Hao YJ, Wei W, Song QX, Chen HW, Zhang YQ, et al. (2011) Soybean NAC
transcription factors promote abiotic stress tolerance and lateral root formation
in transgenic plants. Plant J 68: 302–313.
13. Zhu JK (2008) Epigenome sequencing comes of age. Cell 133: 395–397.
14. Chinnusamy, Gong Z, Zhu JK (2008) ABA-mediated epigenetic processes in
plant development and stress responses. J Integr Plant Biol 50: 1187–1195.
15. Boyko A, Kovalchuk I (2008) Epigenetic control of plant stress response. Environ
Mol Mutagen 49: 61–72.
16. Bilichak A, Ilnystkyy Y, Hollunder J, Kovalchuk I (2012) The progeny of
Arabidopsis thaliana plants exposed to salt exhibit changes in DNA methylation,
histone modifications and gene expression. PLoS One 7(1): e30515.
17. Tariq M, Paszkowski J (2004) DNA and histone methylation in plants. Trends
Genet 20: 244–252.
18. Peters AH, Schubeler D (2005) Methylation of histones: playing memory with
DNA. Curr Opin Cell Biol 17: 230–238.
19. Fuchs J, Demidov D, Houben A, Schubert I (2006) Chromosomal histone
modification patterns – from conservation to diversity. Trends Plant Sci 11: 199–
208.
20. Paszkowsk i J, Whitham SA (2001) Gene silencing and DNA methylation
processes. Curr Opin Plant Biol 4: 123–129.
21. Bird A (2002) DNA methylation patterns and epigenetic memory. Genes &
Development 16: 6–21.
22. Finnegan EJ, Genger RK, Peacock WJ, Dennis ES (1998) DNA methylation in
plants. Annu Rev Plant Physiol Plant Mol Biol 49: 223–247.
23. Zhang XY, Yazaki J, Sundaresan A, Cokus S, Chan SW, et al. (2006) Genome-
wide high-resolution mapping and functional analysis of DNA methylation in
Arabidopsis. Cell 126: 1189–1201.
24. Vaillant I, Paszkowski J (2007) Role of histone and DNA methylation in gene
regulation. Curr Opin Plant Biol 10: 528–533.
25. Lachner M, O’Sullivan RJ, Jenuwein T (2003) An epigenetic road map for
histone lysine methylation. J Cell Sci 116: 2117–2124.
26. Earley K, Lawrence RJ, Pontes O, Reuther R, Enciso AJ, et al. (2006) Erasure of
histone acetylation by Arabidopsis HDA6 mediates large-scale gene silencing in
nucleolar dominance. Genes Dev 20: 1283–1293.
27. Chen ZJ, Tian L (2007) Roles of dynamic and reversible histone acetylation in
plant development and polyploidy. Biochim Biophys Acta 1769: 295–307.
28. Chen LT, Luo M, Wang YY, Wu K (2010) Involvement of Arabidopsis histone
deacetylase HDA6 in ABA and salt stress response. J Exp Bot 61: 3345–3353.
29. Tamaru H, Selker EU (2001) A histone H3 methyltransferase controls DNA
methylation in Neurospora crassa. Nature 414: 277–283.
30. Jackson JP, Johnson L, Jasencakova Z, Zhang X, PerezBurgos L, et al. (2004)
Dimethylation of histone H3 lysine 9 is a critical mark for DNA methylation and
gene silencing in Arabidopsis thaliana. Chromosoma 112: 308–315.
31. Johnson L, Cao X, Jacobsen S (2002) Interplay between Two Epigenetic Marks:
DNA Methylation and Histone H3 Lysine 9 Methylation. Curr Biol 12: 1360–
1367.
Epigenetic Modifications Affect Salt Response
PLoS ONE | www.plosone.org 10 July 2012 | Volume 7 | Issue 7 | e41274
32. Fahrner JA, Eguchi S, Herman JG, Baylin SB (2002) Dependence of histone
modifications and gene expression on DNA hypermethylation in cancer. Cancer
Res 62: 7213–7218.
33. Steward N, Ito M, Yamaguchi Y, Koizumi N, Sano H (2002) Periodic DNA
methylation in maize nucleosomes and demethylation by environmental stress.
J Biol Chem 277: 37741–37746.
34. Choi CS, Sano H (2007) Abiotic-stress induces demethylation and transcrip-
tional activation of a gene encoding a glycerophosphodiesterase-like protein in
tobacco plants. Mol Genet Genomics 277: 589–600.
35. Tsuji H, Saika H, Tsutsumi N, Hirai A, Nakazono M (2006) Dynamic and
reversible changes in histone H3-Lys4 methylation and H3 acetylation occurring
at submergence-inducible genes in rice. Plant Cell Physiol 47: 995–1003.
36. Kim JM, To TK, Ishida J, Morosawa T, Kawashima M, et al. (2008) Alterations
of lysine modifications on histone H3 N-tail under drought stress conditions in
Arabidopsis thaliana. Plant Cell Physiol 49: 1580–1588.
37. Smyth GK, Yang YH, Speed TP (2003) Statistical issues in microarray data
analysis. In: Brownstein MJ, Khodu rsky AB, eds. Functional Genomics:
Methods and Protocols. Volume 224, 111–136 Humana Press, Totowa.
38. Livak KJ, Schmittgen TD (2001) Analysis of relative gene expression data using
real-time quantitative PCR and the 2(-Delta Delta C(T)) Method. Methods 25:
402–408.
39. Jackson JP, Lindroth AM, Cao X, Jacobsen SE (2002) Control of CpNpG DNA
methylation by the KRYPTONITE histone H3 methyltransferase. Nature 416:
556–560.
40. Zhu J, Jeong JC, Zhu Y, Sokolchik I, Miyazaki S, et al. (2007) Involvement of
Arabidopsis HOS15 in histone deacetylation and cold tolerance. Proc Natl Acad
Sci U S A 105, 4945–4950.
41. Agius F, Kapoor A, Zhu JK (2006) Role of the Arabidopsis DNA glycosylase/
lyase ROS1 in active DNA demethylation. Proc Natl Acad Sci U S A 103:
11796–11801.
42. Shibukawa T, Yazawa K, Kikuchi A, Kamada H (2009) Possible involvement of
DNA methylation on expression regulation of carrot LEC1 gene in its 59-
upstream region. Gene 437: 22–31.
43. Pokholok DK, Harbison CT, Levine S, Cole M, Hannett NM, et al. (2005)
Genome-wide map of nucleosome acetylation and methylation in yeast. Cell
122: 517–527.
44. Millar CB, Grunstein M (2006) Genome-wide patterns of histone modifications
in yeast. Nat Rev Mol Cell Biol 7: 657–666.
45. Li X, Wang X, He K, Ma Y, Su N, et al. (2008) High-resolution mapping of
epigenetic modifications of the rice genome uncovers interplay between DNA
methylation, histone methylation, and gene expression. Plant Cell 20: 259–276.
46. Egger G., Liang GN, Aparicio A, Jones PA (2004) Epigenetics in human disease
and prospects for epigenetic therapy. Nature 429: 457–463.
47. Mathieu O, Probst AV, Paszkowski J (2005) Distinct regulation of histone H3
methylation at lysines 27 and 9 by CpG methylation in Arabidopsis. EMBO J
24: 2783–2791.
48. Tariq M, Saze H, Probst AV, Lichota J, Habu Y, et al. (2003) Erasure of CpG
methylation in Arabidopsis alters patterns of histone H3 methylation in
heterochromatin. Proc Natl Acad Sci U S A 100: 8823–8827.
49. Soppe WJ, Jasencakova Z, Houben A, Kakutani T, Meister A, et al. (2002) DNA
methylation controls histone H3 lysine 9 methylation and heterochromatin
assembly in Arabidopsis. EMBO J 21: 6549–6559.
50. Malagnac F, Bartee L, Bender J (2002) An Arabidopsis SET domain protein
required for maintenance but not establishment of DNA methylation. EMBO J
21: 6842–6852.
51. Naumann K, Fischer A, Hofmann I, Krauss V, Phalke S, et al. (2005) Pivotal
role of AtSUVH2 in heterochromatic histone methylation and gene silencing in
Arabidopsis. EMBO J 24: 1418–1429.
52. Xia R, Wang J, Liu C, Wang Y, Zhai J, et al. (2006) ROR1/RPA2A, a putative
replication protein A2, functions in epigenetic gene silencing and in regulation of
meristem development in Arabidopsis. Plant Cell 18: 85–103.
53. Kapoor A, Agarwal M, Andreucci A, Zheng X, Gong Z, et al. (2005) Mutations
in a conserved replication protein suppress transcriptional gene silencing in a
DNA-methylation-independent manner in Arabidopsis. Curr Biol 15: 1912–
1918.
Epigenetic Modifications Affect Salt Response
PLoS ONE | www.plosone.org 11 July 2012 | Volume 7 | Issue 7 | e41274
